QuarterWatch™ (includes data from Quarter 3, 2018) Focus on Three Psychoactive Drugs: Gabapentin, Pregabalin, and Pimavanserin
The latest issue of ISMP’s QuarterWatch™ (see description in box below) focuses on two older but widely used analogs of the inhibitory gamma-aminobutyric acid (GABA) neurotransmitter and a new type of antipsychotic medication approved for Parkinson’s disease psychosis. Key findings include:
Extensive patterns of potentially unsafe use of the GABA analogs, gabapentin (NEURONTIN, others) and pregabalin (LYRICA)
New questions about both the safety and benefits of pimavanserin (NUPLAZID)
The following is a summary of the full QuarterWatch™ report, which can be found, with additional references by clicking here.
In the third quarter of 2018, the US Food and Drug Administration (FDA) received 330,912 new adverse drug event reports involving 1,526 different medications. To capture a broader picture, we examined the most recent 12 months of available FDA Adverse Event Reporting System (FAERS) data, ending with September 2018. During this 1-year period, FDA received 1.4 million adverse event reports; 135,196 of the reports were associated with a fatal outcome. However, 43,781 (32%) of the reports of death included no additional information except that the patient died. Whether the drug contributed to the patient’s death was not determined in these reports. For perspective, 43,781 is a greater number of deaths than annual US fatalities associated with motor vehicle accidents and twice as many as for homicides. Limited information about reported deaths was also an issue in evaluating pimavanserin, as discussed below. This illustrates the need to update FDA’s adverse event reporting regulations and guidances, and to develop protocols for evaluating the possible role of a drug in a patient’s death. Managing the risks of therapeutic drugs requires accurate information about serious injuries and deaths to which a suspect drug may have contributed.
Potentially Unsafe Use of GABA Analog Drugs
Gabapentin is approved to treat postherpetic neuralgia in adults and as adjunctive therapy for some forms of epileptic seizures. It is an analog or synthetic form of GABA, a major inhibitory neurotransmitter. A different formulation (gabapentin enacarbil) is approved for postherpetic neuralgia and restless legs syndrome under the brand HORIZANT. Gabapentin is so widely used for other purposes that it is reported to be taken by more adults (approximately 8.7 million adults take it annually) than any other psychoactive medication except HYDROcodone with acetaminophen.
The other GABA analog, pregabalin, is approved for the same indications as gabapentin as well as for fibromyalgia and neuralgia associated with diabetic neuropathy or spinal cord injury. Pregabalin is less widely used (approximately 1.6 million adults take it annually) and, unlike gabapentin, it is designated as a controlled substance (Schedule V). Pregabalin is only available as a brand product, Lyrica, which costs 40 times more than gabapentin (approximately $460 compared to $11.50 per month).
Using data from a large 2016 survey of medical care in the US,1 we evaluated gabapentin and pregabalin patterns of use. We also analyzed the safety profiles of these two drugs using the most recent 12 months of FAERS data (13,692 total reports) and the literature. Through this analysis, we have identified a pattern of potentially inappropriate and unsafe use so pervasive that addressing it should rank as a major public priority.
Off-label and unapproved use. Both drugs are frequently used off-label to treat other forms of pain, migraine headaches, bipolar disorder, premenstrual syndrome, alcohol withdrawal, and a host of other conditions. Some physicians may use GABA analog drugs “for almost any type of pain,” according to an editorial in the New England Journal of Medicine warning about the growing use.2 Another commentary concluded that up to 95% of current gabapentin use could be for off-label indications.3 The long history of off-label use was partially a result of company marketing campaigns that promoted unapproved uses, which subsequently led to hundreds of millions of dollars in civil and criminal penalties. Investigations connected to this litigation also revealed instances where the scientific literature had been deliberately manipulated to promote increased sales.
Abuse. Another safety concern is outright abuse, either to achieve a euphoric effect or in hopes of potentiating opioids, benzodiazepines, or sedative-hypnotics. In the FAERS data, more than 1,300 reports described withdrawal symptoms, drug abuse, intentional misuse, and overdoses. Hundreds of these reports described fatal overdoses, many of which involved combinations with other drugs known to be abused, notably opioids and anxiolytics. A 2017 survey of overdose deaths in Kentucky showed that 32% had included gabapentin in a cocktail of lethal drugs, most often morphine or fentaNYL.4 Another study of opioid-related deaths in Canada found that 12.3% were also taking gabapentin, which nearly doubled the risk of death compared to those not taking gabapentin.5 A study in Australia showed an increase in overdose and suicide deaths linked to the increasing use of pregabalin.6
Untested concomitant use. GABA analogs are often used concomitantly with other medications, increasing the risk of interactions, overdoses, or inhibition/potentiation of the effects of other needed medications. Our analysis of medication use revealed that half of all patients taking a GABA analog drug were also taking 10 or more other drugs. Specifically, 34.7% of patients taking a GABA analog drug were also sustained users of opioids; 17.8% regularly took another drug that activated GABA receptors (e.g., ALPRAZolam, zolpidem); and 44% regularly took an antidepressant with a GABA analog. Because treatment failure is a common problem with antidepressants, it is possible that GABA analogs are being used off-label on the chance it might increase therapeutic effect. It also appears that antidepressants (mostly tricyclics) and GABA analogs might be combined in hopes of relieving chronic pain.
Other adverse events. In the 12 months of FAERS data, we analyzed more than 2,500 reports signaling that gabapentin or pregabalin was ineffective or had aggravated the condition. Nearly one-third of these reports were associated with an unapproved indication, particularly with gabapentin. We also identified 729 reports of cognitive impairment, including memory loss, memory impairment, confusion, dizziness, and falls in which these drugs were designated as the “primary suspect drug.”
Conclusion. Action is needed to reduce the inappropriate use of GABA analog drugs. Our analysis is consistent with concerns expressed in other scientific forums.3,5,7 FDA should consider reclassifying gabapentin as a controlled substance (some states have already acted to restrict the drug8) and investigate other measures to reduce an overall pattern of unsafe use. Treatment guidelines and physician education are needed to discourage untested use and to increase patient monitoring to ensure prompt discontinuation in cases where the condition is aggravated, interactions occur, or the drug is ineffective.
FDA Reassurance of Pimavanserin Not Warranted
Pimavanserin is approved for treating hallucinations and delusions associated with Parkinson’s disease psychosis. Unlike conventional antipsychotics, its primary effects are on serotonin rather than dopamine receptors. The November 2017 issue of QuarterWatch™ raised concerns about both the safety and benefits of this drug, which were also questioned in FDA’s initial medical review, resulting in an unheeded recommendation to reject the drug.9 Since then, other news media have reported additional safety concerns, notably hundreds of reported deaths in the FAERS data. After these safety concerns were raised in a Congressional hearing, FDA conducted a new safety review of pimavanserin. In September 2018, FDA announced the results of this review in a Drug Safety Communication advising that, “FDA analysis finds no new or unexpected safety risks associated with Nuplazid (pimavanserin).”
We obtained the complete FDA safety reviews, new studies, and the most recent 12 months of FAERS data through September 2018 to conduct a reassessment of pimavanserin. Our results? Rather than being reassured about the safety of the drug, we found weak and incomplete data on patient deaths, new questions about efficacy, and a misleading Boxed Warning in prescribing information that did not adequately inform about the possibility of increased risks and death. Here are the highlights of what we found:
Incomplete death reports. The FDA review focused on 893 reports of death, primarily submitted by the manufacturer, Acadia Pharmaceuticals. Most of the reports had such limited information that FDA safety reviewers could not determine whether pimavanserin contributed to death. Our review of the FAERS data now show 1,339 reported patient deaths since the drug was first approved in 2016 until September 2018, but 522 (39%) cases contained no event description beyond the report term “death.” The company told us that its extensive direct contact with patients and caregivers likely led to the discovery of more cases of death than comparable drugs. The problem is compounded by FDA regulations that require a company to report within 15 days of discovering any patient death, whether or not a drug’s role was suspected, investigated, or confirmed.
More questions about benefits. In pre-approval testing, pimavanserin did not demonstrate a statistically significant benefit in 3 of 4 clinical trials. Concerns about the lack of benefit with this drug were reinforced by new data from actual clinical use cited in FDA’s review. It showed that after just 2 months, 45% of the patients started on pimavanserin had died or discontinued the drug. FDA concluded that the most plausible explanation for the high discontinuation rate had to do with “differences in how patients and their physicians would respond to an apparent lack of efficacy” in the real world as opposed to during a clinical trial. In addition, a large new clinical trial in a different patient population, Alzheimer’s patients with psychosis, failed to show treatment benefit after 12 weeks.
Concerns about Boxed Warning. FDA’s conclusion that no labeling changes were needed for pimavanserin was based in part on its view that even a substantial increase in mortality among patients taking the drug would be consistent with the existing Boxed Warning. The Boxed Warning says: “Increased mortality in elderly patients with dementia-related psychosis. Elderly patients with dementia-related psychosis treated with antipsychotic drugs are at an increased risk of death. NUPLAZID is not approved for the treatment of patients with dementia-related psychosis unrelated to the hallucinations and delusions associated with Parkinson’s disease psychosis.”
The problem is that this warning applies to a different patient population—elderly patients with dementia rather than patients with Parkinson’s disease experiencing hallucinations and delusions. The warning also refers to a different group of drugs, conventional antipsychotics that target dopamine neurotransmitters, and focuses on off-label use of the drug, not that pimavanserin might also increase mortality. A mortality study in Parkinson’s patients showed that using conventional antipsychotics more than doubled the risk of death.10
Conclusion. Our reassessment of the recent FDA analysis provides no new reassurance that the benefits of pimavanserin treatment outweigh its risks. Instead the post-market data and a new study warrant increased concern. FDA should re-evaluate whether evidence exists to show the benefits of pimavanserin outweigh its risks, and if it does, the agency should clarify the Boxed Warning to say plainly that these risks could apply to pimavanserin.